This article compared the accuracy of producing patient-specific cranioplasty implants using four different approaches. Benchmark geometry was designed to represent a cranium and a defect added simulating a craniectomy. An 'ideal' contour reconstruction was calculated and compared against reconstructions resulting from the four approaches -'conventional', 'semi-digital', 'digital - non-automated' and 'digital - semi-automated'. The 'conventional' approach relied on hand carving a reconstruction, turning this into a press tool, and pressing titanium sheet. This approach is common in the UK National Health Service. The 'semi-digital' approach removed the hand-carving element. Both of the 'digital' approaches utilised additive manufacturing to produce the end-use implant. The geometries were designed using a non-specialised computer-aided design software and a semi-automated cranioplasty implant-specific computer-aided design software. It was found that all plates were clinically acceptable and that the digitally designed and additive manufacturing plates were as accurate as the conventional implants. There were no significant differences between the additive manufacturing plates designed using non-specialised computer-aided design software and those designed using the semi-automated tool. The semi-automated software and additive manufacturing production process were capable of producing cranioplasty implants of similar accuracy to multi-purpose software and additive manufacturing, and both were more accurate than handmade implants. The difference was not of clinical significance, demonstrating that the accuracy of additive manufacturing cranioplasty implants meets current best practice.
Keywords: Additive manufacturing; accuracy; co-ordinate measurement; computer-aided design; cranioplasty; implant; rapid prototyping; surgery.